Research map / Label-efficient and robust learning / Domain adaptation and generalization
Both Style and Fog Matter: Cumulative Domain Adaptation for Semantic Foggy Scene Understanding
Xianzheng Ma, Zhixiang Wang, Yacheng Zhan, Yinqiang Zheng, Zheng Wang, Dengxin Dai, Chia-Wen Lin
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Decompose transfer into style then fog adaptation rather than adapting directly (ablation) | CuDA-Net · Foggy Zurich-test | mIoU: 42.49 vs 40.21 | Improved |
| Add style adaptation (ablation) | DeepLab-v2 fog-domain segmentation · Foggy Zurich-test | mIoU: 25.89 -> 39.16 | Improved |
| Add fog adaptation after style adaptation (ablation) | CuDA-Net · Foggy Zurich-test | mIoU: 39.16 -> 42.49; +3.33 mIoU | Improved |
| Add direct dual-factor adaptation after style and fog adaptation (ablation) | CuDA-Net · Foggy Zurich-test | mIoU: 42.49 -> 43.06 | Improved |
| Use cyclical training with T=2 rather than T=1 (ablation) | CuDA-Net without cumulative loss · Foggy Zurich-test | mIoU: 45.32 -> 45.78 | Improved |
| Vary cyclical training from T=1 to T=3 (ablation) | CuDA-Net without cumulative loss · Foggy Zurich-test | mIoU: 45.32; 45.78; 45.45 | No clear effect |
| Use L2 rather than L1 or cosine distance in cumulative loss (ablation) | CuDA-Net · Foggy Zurich-test | mIoU: 47.64; 47.23; 48.21 | Improved |
| Add dense synthetic foggy images to training | CuDA-Net · Foggy Zurich | mIoU: 48.2 -> 49.1; 0.9% | Improved |
| Use two-step adaptation instead of direct adaptation (ablation) | CuDA-Net · ACDC rain validation | mIoU: 46.2 -> 48.5 | Improved |
| Use two-step adaptation instead of direct adaptation (ablation) | CuDA-Net · ACDC snow validation | mIoU: 44.8 -> 47.2 | Improved |
| Select intermediate-domain images manually rather than with a CNN (ablation) | CuDA-Net · Foggy Zurich-test | mIoU: CNN-based: 47.5, 47.5, 47.2; manual: 48.0, 48.2, 48.2 | Improved |
Closest papers by meaning
- Model Adaptation with Synthetic and Real Data for Semantic Dense Foggy Scene Understanding (ECCV 2018 · similarity 0.91)
- Learning Generalized Segmentation for Foggy-Scenes by Bi-directional Wavelet Guidance (AAAI 2024 · similarity 0.89)
- Parsing All Adverse Scenes: Severity-Aware Semantic Segmentation with Mask-Enhanced Cross-Domain Consistency (AAAI 2024 · similarity 0.88)
- Semantic Segmentation in Multiple Adverse Weather Conditions with Domain Knowledge Retention (AAAI 2024 · similarity 0.85)
- Semantic Segmentation on Raindrop Degraded Images Using Two-Stage Dual Teacher-Student Learning (AAAI 2025 · similarity 0.85)
- Every SAM Drop Counts: Embracing Semantic Priors for Multi-Modality Image Fusion and Beyond (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
